The other uni's scientists took our gene construct and put into soybean by taking advantage of the natural plant pathogen Agrobacterium. Agro (as we call it) has the ability to insert it's pathogenic DNA randomly into the DNA of the plant host. This insert can occur once or multiple times. We ordered 3 transformation events and they sent us 6! Nice bang for the buck, though 2 lines were not tested at all due to low seed production. Before I got here some seeds for each line were planted and growing. My first day of my PhD was spent collecting and weighing leaf samples to analyze the plants.
Our inserted gene is nice because it is attached to a reporter gene. A reporter gene produces an easily measured protein that is not found in your organism. The one we use (GUS) is an enzyme that can produce fluorescent molecules which are easily measured on a spec, making the entire process fairly automated. I tested all of the plants for both GUS activity and presence/absence of the gene using PCR. For the most part all the plants tested positive, which is expected in the T1 generation. So now we need to move on to the T2 generation and try to locate a single copy, homozygous line (two or three would be better). To explain what that means, let's step back to genetics 101.
For the genes of sexually reproductive organisms, there are typically 2 copies of the gene: one from Mom and one from Dad. These genes can be the same or they can be different.If they are the same, the organism is said to be homozygote. If they are different there are two types: dominant (the one that gets used) and recessive (the one that doesn't) and the organism is said to be heterozygote. Visually you cannot distinguish between a homozygote dominant and a heterozygote dominant-recessive since they both will use the dominant gene. To tell them apart you have to go to the next generation. This is often visualized by the classic Punnet Square:
|Punnet Square showing two heterozygotes mating, 1 offspring will be dominant (A) homozygote, 1 will be recessive (a) homozygote and 2 will be heterozygotes and appear dominant.|
When you transform an organism you often get either a heterozygote or a hemizygote (only one copy of the gene exists due to it not having a match in the genome). So to achieve a homozygote line, you have to go to the next generation and count them. If you get 3:1 dominant to recessive, as shown in the Punnet Square above, you know that one of them will be your homozygote. To confirm exactly which 1 of the 3 is the homozygote, you have to plant the next generation and only dominant plants grow. Once we get to that stage, I can actually begin testing hypothesis in experiments.
This week I planted the T2 generation of 1 of our 6 lines (50 seeds). In two weeks I will screen them to find those that have the copy of the gene, which if it is present it will be expressed so it is analogous to the dominant recessive in the Punnet Square. Hopefully I will see approximately 38 plants that are dominant and 12 that are recessive which is close to the 3:1 predicted from the mating of our T1 heterozygote generation. If I see a vastly different ratio it probably indicates that the gene was inserted more than once, which makes the plants very difficult to use in experiments. Testing is actually pretty easy, in addition to our gene of interest, the transformed soybeans also carry an herbicide resistant gene. All I have to do is paint one leaf with a little herbicide, wait a week and then score it based on appearance.
The problem with this, of course, is that it takes 2 weeks for them to get large enough to treat without killing the plant and then a week after application to get results. Space limitations as they are means it will take me at least 12 weeks (3 months) to get all the lines tested and the potential plants moved up to the greenhouse where they will take an additional four months before we will have seeds to grow the T3 generation and confirm we have a single insertion homozygous dominant line (all of the offspring will be resistant if they are homozygous). Those seeds will take another 5 months to have seeds that can be experimented on once they are planted. And those are the seeds I can use for experiments, so somewhere between 8 and 11 months from now I will have seeds I can work with from at least one but hopefully 2 or 3 of the lines. Making GMO's is not as easy as a lot of people seem to think.